Die Kuste, 81 (2014), 273-290 
281 
Table 3: Overview of statistical metrics for water level elevation at German stations — N de 
scribes the total number of measurements, a is the standard deviation from mean, RMSD is the 
bias-corrected root mean square deviation from measurements and r the correlation coefficient. 
Water level elevation 
Station 
Observations 
BSH-HBM 
N 
G 
mean 
Bias 
G 
RMSD 
r 
fml 
[ml 
[ml 
[ml 
[ml 
Kiel-Holtenau 
8158 
0.25 
0.05 
0.07 
0.28 
0.12 
0.90 
Koserow 
8301 
0.21 
0.10 
0.18 
0.23 
0.10 
0.90 
Sassnitz 
5792 
0.21 
0.13 
0.15 
0.22 
0.10 
0.89 
Travemuende 
8472 
0.24 
0.07 
0.08 
0.28 
0.12 
0.90 
Wamemuende 
7477 
0.22 
0.08 
0.11 
0.25 
0.11 
0.89 
3.2 Currents 
Due to the very high natural local variability of currents — caused e.g. by local topographic 
effects — on die one hand, and die model immanent spatial averaging on the other hand, 
it is always challenging to do direct model-observation comparisons. In addition to tiiat, 
only a sparse set of measurements is available. For 2008 only data from a few stations in 
the Baltic can be used. A comparison between observed current speed and the current 
speed of die appropriate model cell shows for most part a good agreement, indicated by a 
modelled standard deviation diat is in die same range tiian die observed one at all stations 
and a bias tiiat is lower dian 10 cm/s at most stations. As an example die time series for 
surface- and bottom current velocity at station Arkona is shown in Fig. 7. 
01/01/08 31/03/08 29/06/08 27/09/08 26/12/08 
Date 
01/01/08 
31/03/08 
29/06/08 
Date 
27/09/08 
26/12/08 
Figure 7: Current velocity time series at surface (above) and bottom (below) at Arkona station. 
Observations are shown in black, BSH-HBM data in blue.